CoDro

Protecting critical infrastructures, also in the context of sustainability and efficiency, requires reliable monitoring and surveillance technologies. Operators of supra-regional power grids, for example, have to invest heavily in the maintenance and damage detection of high-voltage lines. In the future, this task could be automated using image recordings from autonomous drones as well as AI-based image analysis and damage classification.

As part of the CoDro (Common AI Algorithms for Autonomous Drones Operation) project, a feasibility study was carried out for 50Hertz Transmission GmbH, one of the four transmission system operators in Germany, in which drone-based and AI-based technologies for monitoring electricity pylons were identified and investigated. The aim was to be able to automatically detect damage to high-voltage lines in the future. To this end, autonomous drones are to take high-resolution static images of the electricity pylons for infrastructure monitoring and condition monitoring. Based on this image data, defects will now be localized and classified using AI algorithms and trained neural networks. In addition, grid conditions, faults, and anomalies are recorded in a database to trigger maintenance as required - for example in the event of large-scale acute corrosion on pylons, conductor cables, and insulators.

Download: Press release

Participating FMD institutes:

Fraunhofer HHI, IZM